In June, GitHub encountered two performance incidents affecting its services. The first incident on June 7 lasted over two hours and was caused by a specific customer activity that led to throttled job processing, resulting in workflow delays and issues with pull requests. The problem was traced back to a large number of pushes to a repository with a particular data shape, which exhausted the worker pool. GitHub resolved the issue by temporarily disabling the source and implementing changes to improve detection and recovery times. The second incident on June 29, lasting 32 minutes, was due to a failover test that inadvertently caused a production outage in parts of North and South America. This test revealed a network configuration issue in the secondary Internet edge facility, which has since been corrected. GitHub is addressing configuration and failover testing gaps to enhance resilience and is committed to ensuring high availability, with improved testing to minimize customer impact.